“Angelo Straffa” Department of Legal Studies

Bocconi Legal Studies Research Paper No. 2433149

(10 June 2013)

INTEGRATING AFRICAN MARKETS INTO THE GLOBAL

EXCHANGE OF SERVICES: A CENTRAL AFRICAN PERSPECTIVE

by

Regis Y. Simo

[Published in the Law and Development Review, Vol. 6, No 2 (2013), pp. 255-297]

This paper can be downloaded without charge from the Social Science Research Network

Electronic Paper Collection: http://ssrn.com/abstract=2433149

1

Integrating African Markets into the Global Exchange of Services: a Central African

Perspective

Regis Y. Simo

Abstract:

Services liberalisation has gradually become very important for growth in developed and less-

developed countries alike and can, as such, be seen as development prospects for sub-Saharan

Africa where numerous economic integration attempts are stories of repeated failures. Despite

the abundant literature on PTAs, however, little attention has been given to Central Africa

Economic and Monetary Community (CEMAC) as a trade bloc. This is an attempt to address

that dearth

At a time when “boosting intra-African trade” is gaining currency on the continent, this article

tests the compatibility of the potential CEMAC economic integration agreement (EIA) against

the background of the existing framework and argues that Central Africa countries would be

in a better position to integrate their economies after widening the borders of their

individually tiny markets. Analysing the legal discipline behind services Preferential Trade

Agreements (PTAs) under the General Agreement on Trade in Services (GATS) and how

CEMAC’s agreement fits into this legal landscape, this article further advocates that this sub-

group of countries should go beyond the Enabling Clause self-contentment and embark on a

deeper (and comprehensive) integration.

Keywords: CEMAC, WTO, Trade in Services, Africa, Preferential Trade Agreements

Contents

I. INTRODUCTION 2

II. GLOBALISATION AND REGIONALISM: A CASE FOR SERVICES 3

III. TRADE LIBERALISATION IN SERVICES: GATS AND AFRICAN

(DEVELOPING) COUNTRIES 6

A. Some Basic Facts about services 6

B. The GATS 7

1. Overview 7

2. Coverage 8

3. Central Africa’s Countries amidst GATS Discipline 8

IV. THE POTENTIAL CENTRAL AFRICA SERVICES ECONOMIC

INTEGRATION AGREEMENT 10

A. Legal Discipline: the Rationale 10

B. Internal Discipline of Central Africa EIA 11

1. Substantial sectoral coverage 11

2. Substantially all discrimination 13

3. Timeframe for liberalisation 14

PhD candidate (International Law and Economics) at Bocconi University (Milan, Italy). Email:

regis.simo@phd.unibocconi.it. Earlier drafts of this paper were presented at both the 2nd

PEPA/SIEL Conference

in Goettingen, Germany (25-26 January 2013) and the Second African International Economic Law Network

Conference in Johannesburg, South Africa (7-8 March 2013).

2

C. External requirements as fortress warning 14

D. Notification 16

E. The Standard of Review 17

F. Flexibility for Developing Countries’ Economic Integration Agreements 19

G. Foreign Direct Investments (FDIs) in EIAs or Liberal Rule of Origin: Article V:6 21

H. Economic Integration Agreements’ Liberalisation Mechanisms 22

1. Scope of Liberalisation and Depth of Commitments 22

2. Liberalisation Modalities 23

V. EFFECTS OF REGIONAL TRADE LIBERALISATION IN SERVICES AND A

POTENTIAL CENTRAL AFRICA SERVICES TRADE BLOC 24

A. CEMAC Common Market Ambitions 25

B. Article V GATS and Central Africa Services Trade Agreement: Coexistence or Deference? 26

1. Effects of Services PTAs 26

2. Some Implications on CEMAC 27

VI. CONCLUSIONS 28

I. INTRODUCTION

Talking about regionalism nowadays may seem old-fashioned because the phenomenon is

neither new to the world trading system nor to Africa. Yet, the 2011 WTO Trade Report

dedicated to Preferential Trade Agreements (PTAs)1 is another indication that everything has

not been said about this area of trade policy where cohabitation and coherence with the

multilateral rules are still making debates.2 In that context, services liberalisation has

gradually become very important for growth in developed and less-developed countries alike3

and can, as such, be seen as development prospects for sub-Saharan Africa where numerous

economic integration attempts are stories of repeated failures. Despite the abundant literature

on PTAs, however, little attention has been given to Central Africa Economic and Monetary

Community (CEMAC) as a trade bloc.4 This is an attempt to address that dearth.

It is also established that recent PTAs between countries in the Northern Hemisphere have

a service component.5 So unsurprisingly is the case of CEMAC that is merely a replica of the

EU success story, even though services chapter has thus far remained marginal. The Doha

1 See World Trade Organization, World Trade Report 2011, The WTO and Preferential Trade Agreements: From

Co-existence to Coherence (Geneva: WTO, 2011). 2 See for instance R. Baldwin, Multilaterising Regionalism: Spaghetti Bowls as Building Blocs on the Path to

Global Free Trade, 29 The World Economy, no.11 (2006), 1451-1518. 3 On the importance of services for economic growth (and the desire to further remove existing barriers to their

trade in the Doha framework and beyond) see B. Hoekman and A. Mattoo, Services Trade Liberalization and

Regulatory Reform: Re-invigorating International Cooperation, World Bank Policy Research Working Paper

No. WPS 5517 (January 2011). See also UNCTAD, Trade in Services and Development Implications (Geneva:

UNCTAD/TD/B/COM.1/85, 2007). 4 An exception being J.T. Gathii, African Regional Trade Agreements as Legal Regimes (Cambridge: Cambridge

University Press, 2011), where the author in Chapter IX discusses CEMAC alongside other monetary unions

such as the West African Economic and Monetary Union (UEMOA) and the West African Monetary Zone

(WAMZ). Note that the study of CEMAC in this paper is without prejudice to the “rationalisation and

harmonisation of African RECs” project that would entail the replacement of the latter by a wider REC – the

Economic Community of Central African States (ECCAS). 5 S. Stephenson, “GATS and Regional Integration”, in P. Sauvé and R. M. Stern (eds.), GATS 2000: New

Directions in Services Trade Liberalization (Washington, D.C.: Brookings Institution Press, 2000), p. 509.

3

Round stalemate, which has seen the rather meek involvement of African countries, in

particular, has also reinvigorated interests of WTO Members to continue trading on a

preferential basis. And today, the challenge of regionalism in Africa lies in its capacity to

build confidence that would promote investment, driver of economic growth.

At a time when “boosting intra-African trade” is gaining currency on the continent,6 it is

expected of the harmonisation of regional economic policies and the implementation of joint

infrastructures projects, especially of (producer) services to enhance CEMAC’s trade with

other countries/sub-regions of the continent, strengthen its participation to the global market

and increase capital flows. Taking advantage of the multilateral framework under Article V of

the General Agreement on Trade in Services (GATS), this paper tests the compatibility of the

potential CEMAC Economic Integration Agreement (EIA) against the background of the

existing framework and argues that Central Africa countries would be in a better position to

integrate their economies after widening the borders of their individually tiny markets.

Analysing the legal discipline behind services PTAs and how CEMAC’s agreement fits into

this legal landscape, this paper further advocates that this sub-group of countries should go

beyond Enabling Clause self-contentment and embark on a deeper (and comprehensive)

integration.

II. GLOBALISATION AND REGIONALISM: A CASE FOR SERVICES

Notwithstanding the history of trade based on goods, services has proven to be the fastest

growing sector, justifying their presence on the Uruguay Round agenda.7 To the exception of

some very few, however, developing countries, and sub-Saharan Africa (SSA) countries in

particular, are still reluctant to grant market access to foreign services and service providers.

Since 2000 and the ensuing launching of the Doha Development Agenda (DDA) they still

have not made noticeable commitments in that sense at the multilateral level.8 And the Doha

deadlock continues to feed Members’ appetite to go for that second-best opportunity.9

Therefore, to posit like Baldwin that “regionalism is here to stay”10

sounds much like a truism

today. The WTO 2011 report further reminds us that despite little novelty in the analysis of

PTAs, there remains a ground to look at the typology of recent waves of regionalism. It is in

6 See for instance Paul Brenton and Gözde Isik (eds.), De-fragmenting Africa: Deepening Regional Trade

Integration in Goods and Services (Washington, D.C.: World Bank, 2012), on the fact that Africa as a whole

trades too little with itself, hence a desire to deepen integration in both goods and services in order to reap the

fruits of trade liberalisation. See also the African Union Action Plan for boosting intra-African trade, available

at: <http://www.au.int/en/sites/default/files/Action%20Plan%20for%20boosting%20intra-

African%20trade%20F-English.pdf>, accessed 15 June 2012. 7 See J. Marchetti and P. C. Mavroidis, The Genesis of the GATS (General Agreement on Trade in Services), 22

European Journal of International Law, no.3 (2011), 689-721, tracing the GATS negotiating history. 8 See e.g. R. Adlung and M. Roy, Turning Hills into Mountains? Current Commitments under the GATS and

Prospects for Change, WTO Staff Working Paper ERSD-2005-01 (March, 2005). 9 There may however be other reasons for this phenomenon of bilateralism. For political reasons why

Governments may prefer regionalism over multilateralism, see e.g. C. Damro, “The Political Economy of

Regional Trade Agreements”, in L. Bartels and F. Ortino (eds.), Regional Trade Agreements and the WTO Legal

System (Oxford: Oxford University Press, 2006), pp. 23-42. Note however that the fact that “regionalism” is on

the rise is no longer a secret. It is gradually appearing not as the “second-best” option portrayed in mainstream

economics but as a fully-fledged policy option. In the words of the Panel in Turkey – Textiles, “regional trade

agreements have greatly increased in number and importance since the establishment of GATT 1947 and today

cover a significant proportion of world trade” (Panel Report, Turkey – Textiles, WT/DS34/R, adopted 19

November 1999, as modified by the Appellate Body Report, WT/DS34/AB/R, at para 9.97). In fact, in the

words of Mavroidis, the status of PTAs has today moved from that of “exception” to that of “right”. See P.

Mavroidis, WTO and PTAs: A Preference for Multilateralism? (or the Dog that Tried to Stop the Bus), 43

Journal of World Trade, no.5 (2010), 1145-1154. 10

See Baldwin (2006), supra note 2.

4

this context that new generations of PTAs are worth analysing and their potential to foster

deeper integration (departing from the “linear” model of integration that has hitherto

characterised African schemes).

Although based on non-discrimination, trade instruments provide exceptions to this core

principle when faced with the rather particular and disadvantaged situations of developing and

least-developed countries (LDCs). This is the essence of special and differential treatment

(SDT) provisions in many WTO Agreements.11

The General Agreement on Trade in Services

(GATS), criticised by some as too intrusive into the national sovereignty of participating

countries, was also hailed by others as the most developing-country-friendly Agreement of the

entire WTO system.12

In addition to the recent services Enabling Clause – the LDC waiver13

–

, the GATS accommodates developing countries and LDCs’ participation concerns in many

respects.14

“Economic Integration” provision of the GATS also allows parties to enter into

PTAs that go contra the obligation not to discriminate, subject to discipline of its Article V.15

Despite the obligation to grant MFN (as the rule) per Article II of the GATS, SSA states

(as less developed countries) have the “right” to enter into economic integration agreements to

liberalise services among and between them if they so desire.16

This in essence is in harmony

with the proliferation of regional economic communities across the continent for more than

half a century now. It is however disappointing how this proliferating and sometimes

overlapping “blocs” have failed for the major part to achieve the objectives ascribed to them

when they were formed. The CEMAC is one of these often recounted failures, critiques

11

SDT provisions, alongside capacity building and technical cooperation, were designed to allow LDCs to

actively participate in the world trade. SDT are of many types. On this score, see E. Kessie, Enforceability of the

Legal Provisions Relating to Special and Differential Treatment under the WTO Agreements, 3 Journal of World

Intellectual Property, no.6 (2000), 955-975, tracing the evolution of developing countries’ negotiations of

GATT/WTO Agreements and concessions accorded to them to accommodate their “weaker” statuses, and the

possible avenues to make these “concessions” enforceable (in the full legal meaning of the word). 12

J. Marchetti and P. Mavroidis, What are the Main Challenges for the GATS Framework? Don’t Talk About

Revolution, 5 European Business Organization Law Review, no. 3 (2004), 511-562, at 513. 13

The LDC waiver was adopted at the 8th WTO Ministerial Conference in Geneva on 17 December 2011 to

enable WTO developing and developed-country Members to provide preferential treatment (hence market

access privileges) to services and service suppliers of LDCs for 15 years from the date of its adoption. Although

it is referred to as “enabling clause”, this terminology should not be confused with the trade in goods scenario

where “enabling clause” refers to a preferential trade agreement among developing countries. In the services

context, this is a “waiver”, providing a legal cover to developing and developed countries when they give

preferential treatment to LDCs (whether WTO Member or not) contrary to Article II of the GATS (on MFN).

Whether this waiver grants LDCs an “actionable” right is another matter altogether. Suffice it to say it does not

in any manner oblige Members to grant preferences. 14

For instance, the Agreement’s Preamble spells out the desire of WTO Members to “facilitate the increasing

participation of developing countries in trade in services and the expansion of their services exports” with

“particular account of serious difficulty of the least-developed countries”. 15

But one caveat is worth making: Article V GATS is not directed exclusively to developing countries, as this

“exclusivity” is essentially dealt with under Article IV GATS (“increasing participation of developing countries”

in the world trade). Rather, Article V GATS offers flexibility (vis-à-vis the multilateral rule) when an agreement

of the type envisaged by that provision has one or more developing countries as its members. That is different

from its GATT counterpart (article XXIV) that designed no special rules concerning PTAs between developed

and developing countries (although goods PTAs between two or more developing countries are dealt with under

the “Enabling Clause”). Additionally, in the process of “progressive liberalisation”, Article XIX.2 GATS also

offers “flexibility” in the manner it will be conducted in this group of countries (progressively opening fewer

sectors, liberalising fewer types of transaction, etc. in line with Article IV objectives). 16

On the status of PTAs from that of “exception” to that of “right”, see Mavroidis (2010), supra note 9.

5

stemming from the fact that African PTAs should primarily – if not solely –be based on trade

integration like other PTAs against which they are compared.17

As of January 2012, around 105 economic integration agreements (EIAs) in the sense of

Article V of the GATS out of about 232 regional trade agreements (RTAs) are in force.18

CEMAC like many RTAs in the developing world has been notified to the WTO under the

“Enabling Clause”19

over goods. Although some of these schemes envisage “service”

liberalisation, that component is yet to be notified to the WTO for the corresponding

discipline to kick in. What’s more, agreements under Article V GATS involving developing

countries amount to half of notifications since 2009, but just a handful of them actually

involve African countries.20

This paper envisages the multilateral framework serving as a benchmark for a better

regional integration in central Africa. It is a fact to say that the recent proliferation of PTAs

has not always been conditioned, if at all, on a prior satisfaction of the relevant legal regime.21

We proceed to ask ourselves whether the situation at some point need not be reversed for the

multilateral discipline to serve as the benchmark for a better intra-regional trade.22

Since

obstacles to integration often persist despite the proclaimed intentions, it could be that

reliance on GATS V and its requirements would boost intra-regional trade, which in turn

would serve its purpose as building block to the wider multilateral liberalisation.23

Therefore,

by respecting the existing legal regime, these countries can deviate24

from MFN in a more

efficient manner. But it should be remembered that trade liberalisation and integration into the

world economy are not ends in themselves if other factors like geography, resources

endowments, the quality of a country’s institutions and its regulatory framework are not put to

17

On the criticism of the general tendency of overstating the failure of African regional trade agreements

because they usually serve other purposes (apart from trade integration), see J. Gathii, African Regional Trade

Agreements as Flexible Legal Regimes, 35 North Carolina Journal of International Commercial Regulation, no. 3

(2010), 571-668, for whom recounting the so-called “failures” over and over again may be too pessimistic a take

on African regionalism. 18

See Regional Trade Agreements database, available at: <

http://rtais.wto.org/UI/PublicSearchByCrResult.aspx>, accessed 15 June 2012. 19

The Enabling Clause stands for the General Agreement on Tariffs and Trade (GATT) Contracting Parties

“Decision on Differential and More Favourable Treatment, Reciprocity and Fuller Participation of Developing

Countries” of 28 November 1979 (GATT L/4903), GATT BISD 26th Supp. (Geneva, 1980), p.203. It was later

incorporated in the corpus of WTO law by GATT 1994, para. 1(b) (iv). 20

M. Roy, Services Commitments in Preferential Trade Agreements: Surveying the Empirical Landscape,

NCCR-Trade Working Paper No 2012/02 (January 2012). 21

P. Mavroidis, If I Don’t Do It, Somebody Else Will (Or Won’t): Testing the Compliance of Preferential Trade

Agreements With the Multilateral Rules, 40 Journal of World Trade, no.2 (2006), 187-214. 22

But let’s not ignore that non-compliance by a PTA with the multilateral rules does not impact upon its validity

between the Parties who signed it. A PTA is an international treaty on its own and there is consequently no

deference of one to the other since they both stand on an equal footing in international law. Thus, “WTO

Agreements per se have no legal supremacy over Economic Integration Agreements”: T. Cottier and M.

Molinuevo, “Article V GATS”, in R. Wolfrum, P. T. Stoll and C. Feinäugle (eds.), Max Planck Commentaries

on World Trade Law: WTO – Trade in Services, vol. 6 (The Hague: Martinus Nijhoff, 2008), p. 128. But the

undesirable effects might advocate for a PTA-compliant rather than a PTA-rebellious. 23

Since it is less doubt that Baldwin’s concept of “multilateralising regionalism” commends that multilateralism

and regionalism strengthen each another. See F. Söderbaum, Unlocking the Relationship Between the WTO &

Regional Integration Arrangements (RIAs), 35 Review of African Political Economy, no.118 (2008), 629-644, at

630. On how to do that, see R. Baldwin and P. Thornton, Multilateralising Regionalism: Ideas for a WTO Action

Plan on Regionalism (London: Centre for Economic Policy Research, 2010). 24

At least for the time needed to reverse the diverting effects it they exist and which, unfortunately (for the

multilateral regime), might even take longer in services context given its particular nature (bound to be

regulated) and the speed at which multilateral negotiations are being conducted. But we will see later that

services PTAs do not necessarily lead to diversion.

6

contribution.25

All expectations for economic growth and sustainable development do not

therefore have to be placed on a subset of trade policy alone.26

Yet, efforts must be put

together so as not to annihilate the potentials that such subset can contribute for the overall

growth aspirations. And because not so much has been (or likely to be) achieved since the

launching of the Doha Round in terms of commitments, the “dialectic process of world trade

liberalisation stimulated by regional process,”27

is consequently called to take place again, in

the sector of services.

III. TRADE LIBERALISATION IN SERVICES: GATS AND AFRICAN

(DEVELOPING) COUNTRIES

A. Some Basic Facts about services

Liberalisation of services finds its rationale in their role to economic activities at large. In so

far as they are themselves tradable, they constitute for some of them inputs for the trading of

goods and other services. Financial, transportation and other infrastructure services are the

most oft-cited “producer services”, absent which development of trade in goods will be close

to nil.

Services, the fastest growing sector of the world economy, represent two thirds of global

output but contrasted by its share of about 20% of the global trade (“only”).28

They account

for more than two thirds of the Gross National Product (GNP) of the Organisation for

Economic Co-operation and Development (OECD) countries29

and between 60 and 75 percent

of Gross Domestic Product (GDP) and employment in these countries.30

Moreover, services

exports (consisting of mainly tourism and travel services) in developing countries grew about

3 percent rapidly per annum (on a balance-of-payment basis) than developed countries’

exports between 1990 and 2000.31

They account for some 52 percent of developing countries’

GDP and about 35 percent employment.32

The growing importance of this sector on world trade therefore prompted negotiations for

an adoption of a legal instrument at multilateral level, pushed by developed countries chief

among which were the United States and the European Community.33

The Uruguay round

culminated in the adoption of the GATS in 1995 as the first comprehensive and only

25

See J. Marchetti, “Developing Countries in the WTO Services Negotiations: Doing Enough?”, in G. Berman

and P. Mavoidis (eds), WTO Law and Developing Countries (Cambridge: Cambridge University Press, 2007), p.

83. 26

Ibid. 27

T. Cottier, The Challenge of Regionalization and Preferential Relations in World Trade Law and Policy, 1

European Foreign Affairs Review, no.2 (1996), 147-167, at 156. 28

B. Hoekman, and P. Sauvé, “Regional and Multilateral Liberalization of Service Markets: Complements or

Substitutes?, 32 Journal of Common Market Studies no.3 (1994), 282-318, at 284, footnote 1. 29

M. Matsushita, T. Schoenbaum and P. Mavroidis, The World Trade Organization: Law, Practice and Policy

(2nd ed., Oxford: Oxford University Press, 2006), p. 607. See also Wolfrum, Stoll and Feinäugle (eds.) (2008),

supra note 22, p. (ix), where global services exports amounted to 2.6 trillion US Dollar in 2006. 30

Hoekman and Sauvé (1994), supra note 28, at 284. 31

World Trade Organization, A Handbook on the GATS Agreement, (Cambridge: Cambridge University Press,

2005), p. 3. 32

UNCTAD (2007), supra note 3, p. 2. These figures notwithstanding, Africa represents only 10 percent of

services exports in the developing world, see Id, p. 3. 33

The issue was nevertheless already present on the agenda of the Tokyo Round (1973-1979) at the initiative of

the US. See Marchetti and Mavroidis (2011), supra note 7. See also C. Fuchs, “GATS Negotiating History”, in

Wolfrum, Stoll and Feinäugle (eds.) (2008), supra note 22, p. 3.

7

multilateral agreement on services trade, which before then was conducted, when it did, on

bilateral and regional bases.34

Whether the advent of this instrument has increased the flow of

services beyond the pre-GATS level is an empiric question significantly dependent on the

conclusion of Doha (or any other subsequent round of services negotiations) negotiations.35

B. The GATS

1. Overview

Like the GATT for goods, the GATS imposes a discipline on trading parties in services but as

an instrument of “progressive liberalisation”.36

If trade liberalisation in general requires

granting market access to foreign companies by lowering entry barriers to trade, the issue is

not less complex when it comes to services where barriers are not limited to border measures.

GATS, therefore, does not pretend to liberalise at “one shot”, but rather to gradually remove

regulations the purpose and/or effects of which restrict worldwide flow of services – i.e.

“unnecessary regulations”. Nevertheless, it is up to different countries to choose sectors they

wish to liberalise for which they commit themselves (somehow irreversibly).37

The GATS opted for a “flexible” regulatory framework – in many respects38

– whereby

each party’s level of commitment would be decided and clarified ex ante.39

(Domestic)

“Regulations” are the medium through which countries erect barriers to trade in services, for

the large share of trade in services take place domestically, i.e. inside one’s own country. In

order to balance between states’ regulatory objectives and international trade liberalisation,40

GATS creates two categories of obligations: general obligations and specific commitments.

General obligations on the one hand cannot be deviated from (or contracted out of). They

are the general discipline which all parties, by joining the WTO, agree to respect. They are

binding even when one has not undertaken an obligation to liberalise a specific sector. The so-

called “most-favoured nation” (MFN) principle41

falls in this category. Specific commitments

on the other hand bind a member only to the extent it has expressly entered a commitment to

liberalise a particular sector, or a particular mode of supply – (which is a kind of bottom-up

approach). Specific commitments are considered to be the main tools of liberalisation of

GATS in that provisions contained therein aim at limiting the use of certain quantitative

34

Matsushita, Schoenbaum and Mavroidis (2006), supra note 29, p.604. 35

See J. Marchetti and P. Mavroidis (2004), supra note 12, at 523-524. 36

See recital 2 of GATS Preamble. 37

Although Article XXI GATS allows a Member to modify a commitment in its Schedule (after three years), it

is rather difficult to think of how that might happen in practice given that the modifying Member is required to

negotiate “compensatory adjustments” with “affected Members” on an MFN basis, which the latter must agree

with. This is a difficult result to achieve as it involves reaching consensus (by all WTO Members) to allow the

modifying Member to deprive other Members of the advantages they have been enjoying in trading when that

commitment was in force. This particular argument can be advanced on Article V:5 (modification and

withdrawal of commitment in an economic integration agreement). 38

The approach taken by negotiators was to opt for a “progressive liberalisation” for all Members, unresolved

issues to be sorted out during subsequent rounds; and flexibility with respect to the participation of developing

countries and loosened discipline when it comes to services of interest to them, etc. 39

Matsushita, Schoenbaum and Mavroidis (2006), supra note 29, p. 605. 40

It is argued, however, that GATS favours liberalisation over allowing domestic regulation due to the “right of

Members to regulate, and to introduce new regulations, on the supply of services within their territories in order

to meet national policies objectives”, as found in Recital 4 of the Preamble of GATS which, if it were to be

accorded greater impact, should have simply been made part of the provisions of Article VI (Domestic

Regulation). See H. Hestermeyer, “Preamble General Agreement on Trade in Services”, in Wolfrum, Stoll and

Feinäugle (eds.) (2008), supra note 22, p. 27. 41

Article II GATS.

8

restrictions to the provision of services once a party has undertaken to open up a service

sector.42

These disciplines are found in “market access” (Article XVI) and “national

treatment” (Article XVII) provisions.

2. Coverage

The GATS applies to measures by Members affecting trade in services.43

“Measures by

Members” are those taken by central, regional or local governments as well as non-

governmental bodies exercising delegated powers.44

Concerning “services”, the GATS does

not define them proper. Rather, it categorises those to which it applies (to the exclusion of

some), as including any service in any sector except services ‘supplied in the exercise of

governmental authority’.45

This rather wide definition has the merits of embracing some

areas that had in the past remained outside WTO foresight.

Services in the GATS context can be traded in four different ways – or modes of supply.46

We will see later that these modes of supply are of particular importance to regional

integration agreements because of their “substantial sectoral coverage” requirement.47

It is

also worth mentioning that Mode 3, the apparent dominant mode of supply for all sectors

(except transport and tourism services),48

is akin to an international agreement to liberalise

investment in the sense that the opening of a sector amounts to opening up the sector to

foreign investment.49

Mode 1 is the second most important mode of supply, while Mode 4 the

least significant across all sectors, and Mode 2 the mode par excellence for tourism service

trade.50

Overall, if Mode 1 and Mode 2 do pose less problems since they are analogous to the

cross-border trade in goods, the GATS has succeeded to break new grounds with Modes 3 and

4 in establishing multilateral rules that guarantee the opportunities for legal and natural

persons to establish themselves in a foreign market.51

Although Mode 3 requires the

establishment of a foreign supplier firm it does not necessarily imply the presence of staff

with foreign nationalities. It follows that when a foreign establishment elects to employ a

foreign manager for instance, the supply of service is covered by both Mode 3 and Mode 4.

Mode 4, which is the presence of natural person of a foreign nationality in another country to

provide service, can also be found absent commercial presence, because the GATS embraces

the possibility of providing services by individuals in an independent capacity.

3. Central Africa’s Countries amidst GATS Discipline

42

P. Delimatsis and M. Molinuevo, “Specific Commitments: Article XVI GATS”, in Wolfrum, Stoll and

Feinäugle (eds.) (2008), supra note 22, p. 369. 43

Article I:1 GATS. 44

Article I:3(a) GATS. 45

Article I:3(b) GATS. Article I:3(c) GATS on its part defines a service “supplied in the exercise of

governmental authority” as one supplied neither on a commercial basis, nor in competition with one or more

service suppliers, i.e. not-for-profit activities like social security or central banking services. 46

Pursuant to Article I:2 GATS, services can be traded in the following manners: (i) Mode 1 – cross-border

supply; (ii) Mode 2 – consumption abroad; (iii) Mode 3 – commercial presence; (iv) Mode 4 – presence of

natural persons. 47

Article V:1(a) GATS. See Section V. 48

J. Hodge, “Liberalization of Trade in Services in Developing Countries”, in B. Hoekman, A. Mattoo, and P.

English (eds.), Development, Trade and the WTO: A Handbook (Washington, D.C.: The World Bank, 2002), p.

222. See also Marchetti (2007), supra note 25, p. 97. 49

Matsushita, Schoenbaum and Mavroidis (2006) supra note 29, p. 617. 50

Hodge (2002), supra note 48, p. 222. 51

World Trade Organization, Guide to the Uruguay Round Agreements (The Hague: Kluwer Law International,

1999), p. 164.

9

The majority of WTO Members (about four fifth) are not developed countries. But it is rather

controversial the way discussions over services have moved during the Uruguay Round and

the directions taken during DDA. If the majority of developing countries were vehemently

opposed to the introduction of services in the former Round, especially Mode 3 supply, the

tendency was modestly toward the opposite direction in the DDA. In fact, having failed for

some of them (especially Argentina and India) to approach negotiations in a more concerted

manner in the heterogeneous “Café-au-Lait” Group (composed of both developed and

developing countries), the tune soon changed, especially for India who became one of the co-

authors of Mode 4 on the movement of natural persons.52

It also appears that Mode 3 happens

to be the mode of supply currently allowing wide regulatory manoeuvres by host countries

wishing to control foreign companies’ establishments and activities in their markets.

Furthermore, recent developments in, and access to, information technologies by developing

countries have somehow increased their eagerness to see more commitments over Mode 1 (to

which they were also opposed during Uruguay Round for lack of capacities).53

If the above statements may hold true for some developing countries, it is not necessarily

the case for all of them. African countries, not to speak of developing countries as a whole,

are heterogeneous in nature, and many SSA countries do not enjoy the same development

pace compared to the most advanced ones such as China, India, Brazil, Egypt or South

Africa.54

In this environment, stakes are not similar and levels (and willingness) of

commitments uneven. Current multilateral negotiations exhibit these trends. Some developing

countries (mostly middle-income economies) have undertaken commitments comparable and

sometimes more ambitious in depth and breadth to those of developed economies thereby not

relying too much on flexibilities,55

while the poorest among them are characterised by hesitant

stances (sometimes even failing to make use of flexibilities), criticising the process for not

taking into account sectors and supply modes of potential benefit to them.56

The number of

committed sectors by WTO Members in July 2000 just pictured this inequality.57

It then seems logical to infer from this conduct homogeneous among poorer developing

countries – although it is difficult to single out central African countries’ approach –that they

have been cautious in embracing negotiations to open their services markets to competition

from abroad. Even though such moves might bear several meanings in terms of tactics for

future multilateral and regional negotiations or in terms of just a lack of interest, it

nevertheless signals their intentions – past, present and future – to commit or not to.58

Needless to note that protection, whatever the form, is harmful both to the local economy at

large and to domestic consumers in particular on whom the burden of higher barriers to entry

52

Marchetti and Mavroidis (2011), supra note 7. 53

Which Mode became one of the most important challenges for developing countries in the DDA of services

negotiations. See Marchetti (2007), supra note 25, pp. 97, 115-116. 54

The status of some of which is rather controversial since WTO does not have a definition of “developing

countries” on its own, letting each of them to self-elect its status on accession. There is indeed no international

consensus on the concept. 55

R. Adlung et al, “The GATS: Key Features and Sectors”, in Hoekman, Mattoo and English (eds.) (2002),

supra note 45, p. 261. 56

A. Mukerji, Developing Countries and the WTO; Issues of Implementation, 34 Journal of World Trade, no.6

(2000), 33-74, at 33, 39-40 cited by Matsutshita, Shoenbaum and Mavroidis (2006), supra note 29, p. 782.

Enhanced market access is the bone of contention between negotiating actors and the politically sensitive Mode

4 is of utmost importance to developing countries that want to see developed nations commit while in turn

refraining to open further on this mode of supply. Offers as of the end of 2005 has shown this dangling pattern.

On this score, see J. Marchetti (2007), supra note 25, pp.100, 110-114. 57

See Adlung et al (2002), supra note 55, at 263 (Table 27.3). See also Marchetti and Mavroidis (2004), supra

note 12, at 521 and 558 (Table I). 58

Marchetti (2007), supra note 25, p. 90.

10

and of the absence of competition falls in terms of monopolistic prices and sometimes low-

quality goods and services. That is why substantial evidence has been advanced to

demonstrate the welfare reducing effects of policies purporting to limit competition in

services industries, especially producer services, which is a rather frequent pattern in

developing countries as a whole.59

It remains a fact that SSA industries for the most part lack the degree of competitiveness

against developed nations’ firms. It is also no doubt that developing countries have the right,

as given by GATS, to only liberalise at a pace convenient for their development goals and in

sectors of interest to them. Moreover, flexibilities as far as (preferential) trade agreements

have been granted, with the expectation that advantage would be driven from there. If they

then seem not too enthusiastic thus far to commit further at the multilateral table, could

regional agreements not help out? Given that their markets taken individually and even

collectively are for the major part too small as mentioned, exploiting opportunities at regional

levels would then be a step towards wider liberalisation. Because protection is costly, it will

still be more beneficial to end up with more multilateral commitments at some point. And on

this agenda, regional policy choices and individual States’ domestic regulations would be the

benchmarks by which integration will be judged as welfare enhancing enough or not. An

interest for services talks at regional level is not absent though. Discussions within the ACP

group (in which CEMAC countries participate) with the EU is one example.60

Still, what is

the level of liberalisation at central Africa sub-regional level remains an issue. What then

could be feasible to improve upon the stagnating status quo?

IV. THE POTENTIAL CENTRAL AFRICA SERVICES ECONOMIC

INTEGRATION AGREEMENT

A. Legal Discipline: the Rationale

The Panel in Canada – Autos stated that “Article V provides legal coverage for measures

taken pursuant to economic integration agreements, which would otherwise be inconsistent

with the MFN obligation in Article II”.61

As such, WTO Members are allowed to enter into an

agreement to further liberalise trade in services with other Members that accept to be parties

to it. Simply put, Article V as intended by its chapeau may justify the adoption of a measure

inconsistent with certain provisions of GATS provided such measure satisfies the

requirements therein specified as discussed at length below.62

What the law denotes is that for a PTA under GATS to be consistent with the terms of

Article V, thereby complying with the multilateral rules, it must not breach any term of the

59

Ibid, pp. 87-89. Marchetti shows how welfare has been enhanced in those countries whose telecommunication

industries were liberalised, allowing for extended internet penetration that in turn affected in a positive manner

how other services are being traded. The same holds true for maritime transport services liberalisation (in Chile)

in the reduction of transportation costs, and also in distribution services (especially of agricultural products)

where Zimbabwean farmers for instance have seen an increase in their income when competition was introduced

(eliminating monopsony) in the market for the purchase of their production. 60

See Joseph J.L. Correa, L’OMC à l’épreuve des Accords de Partenariat Economique et de l’intégration

économique africaine (Zurich: Schulthess, 2007), pp.144-151. 61

Panel Report, Canada – Certain Measures Affecting the Automotive Industry, WT/DS139/R and WT/DS142/R

para. 10.271. [hereinafter Canada – Autos] 62

Appellate Body Report, Turkey – Restrictions on Imports of Textile and Clothing Products, WT/DS34/AB/R,

para.45 (hereinafter Turkey – Textiles) cited in Cottier and Molinuevo (2008), supra note 22, p. 128.

11

provision that are all mandatory in nature.63

Just like a PTA under GATT, a potential central

Africa services EIA would have to pass both a) the internal and b) the external test, which are

of substantive nature, and c) as a matter of procedure, will have to be notified. GATS being a

ground-breaking instrument with no precedent, interpretation of these requirements are

obviously not crystal clear.

B. Internal Discipline of Central Africa EIA

Satisfying the internal discipline requires an agreement to form an EIA to have (i) “substantial

sectoral coverage”, and (ii) devoid of “substantially all discrimination” except for what is

necessary under Articles XI, XII, XIV and XIV bis, and (iii) achieved within the prescribed

timeframe.

1. Substantial sectoral coverage

In view of limiting perversion of MFN, hence of the world trade as a whole, by concluding

agreements on relatively few sectors and supply modes of interest to them (i.e. PTAs à la

carte), GATS’s discipline requires EIAs to have substantial coverage between its parties.64

For an agreement to have “substantial sectoral coverage” in the sense of Article V:1(a) GATS,

footnote 1 provides for both a quantitative and a qualitative test and two arms with regards to

“sectors” and “modes of supply” in stating that “this condition is understood in terms of

number of sectors, volume of trade affected and modes of supply. [And] in order to meet this

condition, agreements should not provide for the a priori exclusion of any mode of supply.”

An agreement that excludes any mode of supply is a priori disqualified as an EIA as intended

by the GATS. Grey areas persist as to the exact reach of these tests.

With the “qualitative” test pertaining to sectors for instance, questions arise as to whether

the agreement can remain valid after individual services, sub-sector(s) or whole sector(s) have

been excluded. Inferring from the Appellate Body’s decision in Turkey – Textiles where it was

held that “substantially all trade” is not the same as “all trade” but should at least be

“something considerable more than merely some of the trade”, Cottier and Molinuevo submit

that Article V of the GATS does not require all sectors to be covered.65

Rather, the EIA must

not exclude “more than a very limited number of sectors” (i.e. not more than one sector

entirely).66

And a further controversial point is that in assessing the impact of the excluded

sector, consideration has to be had on its economic importance in terms of world trade. In

other words an agreement shall not exclude a “major sector”.67

At first instance, it should not be expected of an EIA to meet higher standard than that

expressed in the GATS itself. In this vein, it would not be at odd with GATS V to exclude air

transport services, maritime transport services, financial services each of which is the subject

of a GATS “Annex” on its own. However, since the purpose of an EIA is generally to obtain

at regional level what is unobtainable at multilateral level, EIA would not pass the test of

GATS V if liberalisation were to be less in coverage. This argument militates for the view that

63

However, although it will make the agreement unlawful, non-compliance with one or more of GATS

provisions does not render the PTA null and void. The issue would then be for interested (and/or third) parties to

issue an MFN claim or ask for (adequate) compensation by way of dispute settlement. See Cottier (1996), supra

note 27, pp.162-163. 64

Cottier (1996), supra note 27, at 158. 65

Cottier and Molinuevo (2008), supra note 22, p.131. 66

Ibid. 67

See the Committee of Regional Trade Agreements Note on the Meetings of 29-30 April and 3 May 1999,

WT/REG/M/22, 4 June 1999, cited in Cottier and Molinuevo (2008), supra note 22, p. 131.

12

EIAs in principle are expected to liberalise further than what actually happens at the WTO’s

table.68

Furthermore, it would rather be awkward to exclude the so-called “producer services”

that are essential for the supply of other services (e.g. financial services), and sometimes of

merchandises themselves (e.g. transport services), especially in SSA where integration is

highly needed to connect inland territories to the rest of the world. Views are nevertheless

divergent on this point since there is no such obligation in GATS texts not to carve out one of

these sectors.69

Services supplied in the exercise of government authority are per se excluded

from GATS’s purview. These controversial points do not only have the power to render

conclusion of EIAs difficult, but also the merit of protecting the MFN principle (which it

however set out to deviate from).70

As for the “quantitative” test, again as far as sectors are concerned, the agreement must

not allow for the exclusion of the sectors which amount to substantial trade between the

parties. This applies to both actual (current) and/or potential trade. If restrictions on particular

services or sectors are to be maintained, therefore, they must not be on those where significant

trade between parties occurs or would occur (in absence of such restrictions).71

Owing to the

difficulties to quantifying the “volume of trade” (actual or potential) for lack of precise data

on international trade in services, assessment of an EIA as WTO compatible or not leans in

favour of the qualitative test.72

Concerning modes of supply that form the second arm of the internal requirement,

footnote 1 to Article V:1(a) provides for the a priori non-exclusion of any mode. Again what

this may mean in reality is far from certain. Cottier and Molinuevo opine that an EIA may

provide for differential degrees of liberalisation for different modes of supply as long as none

of them is excluded entirely.73

And in particular, no EIA should a priori exclude investment

(mode 3) or labour mobility (mode 4).74

But the question is far from settled since it is not

clear whether a requirement by a host country to establish oneself locally (mode 3) before

trading its services (in a very particular sector) in the local market – thereby excluding cross-

border supply (mode 1) – could amount to a violation of the provision.75

Equally unclear is

whether, in trading services under mode 4, all labour mobility must be included in an EIA.

For CEMAC whose aim is to form a customs union, it is important to note that

commitments on the movement of natural persons should not be narrower in scope than under

the GATS, and also that it might be expected (or at least desirable) of its Members to go a bit

further than what is achievable under the GATS (even though GATS V does not actually

differentiate between CUs and FTAs).76

It would then seem at odd with the idea of “regional” trade liberalisation where the idea is

to achieve a high stage of integration to exclude from the outset one mode of supply. This is

68

But difficulties in interpretation resurface if no commitments on MFN basis have been made at all. This is due

to the fact that it is difficult to depict free trade scenario in services since almost all barriers are regulatory in

nature with no obligation to harmonise laws. 69

Cottier and Molinuevo (2008), supra note 22, at 132. 70

Ibid. 71

Ibid., at 132-33. 72

Ibid., at 133. 73

Ibid. 74

P. Van den Bossche, The Law and Policy of the World Trade Organization, (2nd ed., Cambridge: Cambridge

University Press, 2008), p. 711. 75

Stephenson (2000), supra note 5, p. 516. 76

Ibid., footnote 6. However, it may be relevant while reviewing the agreement under Article V:2 GATS.

13

because “the purpose of Article V is to allow for ambitious liberalisation to take place at

regional level, while at the same time guarding against undermining the MFN obligation by

engaging in minor preferential arrangements.”77

As Cottier and Molinuevo summarise:

While a) no mode of supply should be a priori excluded from the agreement, b)

liberalisation commitments may be undertaken with regard to some modes of supply

more than others, provided that c) the lack of commitments with regard to one or

modes of supply does not impair the liberalisation of substantial volumes of trade.78

2. Substantially all discrimination

While featuring a substantial sectoral coverage, an EIA is required by Article V:1(b) to

provide for the absence or elimination of substantially all discrimination between or among

the parties to the agreement. In order words, national treatment in the sense of Article XVII of

GATS must be extended to services and service suppliers of other parties to the agreement in

the same manner as domestic services and service suppliers. This requirement to treat no less

favourably service suppliers of other parties than domestic service suppliers aims at bringing

about level playing fields for undistorted competition, and it must be done in respective

parties’ markets on an equal footing (i.e. on a reciprocal manner).79

Furthermore, Article

V:1(b) should be seen as granting MFN too when national treatment is satisfied. This is

because it is deemed inconsistent with that article if more favourable treatment is to be

accorded to services and service suppliers of one party and not to those of another party to the

agreement,80

which is obviously the main reason why parties conclude PTAs.

The extent to which discrimination can be allowed to remain in such an agreement is not

clear though. Because when this provision requires the elimination of “substantially all”

discrimination, it does not require the elimination of “all” discrimination.81

This entails that

discrimination between nationals and foreigners can be maintained to the extent that it is

necessary under GATS Articles XI (Payments and Transfers), XII (Restrictions to Safeguard

the Balance of Payments), XIV (General Exceptions on: health, safety, taxation, public order,

etc.) and XIVbis (Security exception). These exceptions are provided in order to balance

between the individual States’ legitimate regulatory objectives and those of comprehensive

liberalisation in an EIA. These domestic policy objectives should not be undermined.

However, because this list is not exhaustive, doubts persist as to whether parties to an EIA

are not allowed to discriminate against services suppliers of another member of the agreement

in granting licences for professional services (since Article VII on Recognition is not

specifically exempted) or in granting domestic subsidies (Article XV) and government

procurement of services (Article XIII) nor even applying Emergency Safeguard Measures

(Article X) to their agreement.82

As to whether the requirement to eliminate substantially all discrimination covers both

present and future measures, or whether it means either present or future (i.e. in the

alternative), it has been argued that they are options (or strategies) with regards to the

sector(s) being liberalised for which alternatives to be freely chosen by parties are not

77

Panel Report, Canada – Autos, para. 10.271. 78

Cottier and Molinuevo (2008), supra note 22, at 134. 79

Ibid., at 135. 80

Van den Bossche (2008), supra note 74, p. 712. 81

Ibid., p. 664. See also Stephenson (2000), supra note 5, p. 517. 82

Van den Bossche (2008), supra note 74, p. 664 & Stephenson (2000), supra note 5, p. 517.

14

allowed.83

Article V:1(b) provides for the “(i) elimination of existing discriminatory measures

(a rollback mechanism), and/or (ii) prohibition of new or more discriminatory measures”.

While (i) entails an obligation to liberalise, (ii) prevents parties from introducing new

restrictive measures (a stand-still obligation).84

As posited, these two tracks are simply means

to achieve the main obligation of an EIA to eliminate “substantially all discrimination”.85

A

party would therefore not discharge its obligation neither by removing existing discrimination

prior to the conclusion of the EIA while introducing new such measures afterwards, nor by

refraining from the latter while keeping old discriminating measures in place.

It nevertheless remains a fact that it is still difficult to understand clearly what

“substantially all discrimination” in the sense of Article XVII means because the paucity of

case law under that same provision does not help to clarify what “discrimination” means in

the first place. This is another grey area in GATS’s law.

3. Timeframe for liberalisation

The requirements to liberalise substantially all trade and to eliminate substantially all

discrimination must not necessarily be met immediately, but could be achieved over a

reasonable period of time. According to the provisions of Article V:1(b) in its last paragraph,

these requirements should be achieved “either at the entry of that agreement or on the basis of

a reasonable time-frame”. What “reasonable timeframe” may mean is subject to

interpretation. For want of a discussion of this phrase in the sphere of services, the

Understanding on the Interpretation of Article XXIV GATT 1994 where ‘reasonable

timeframe’ is provided as being not more than ten years, can be offered as guideline,86

although some Members suggested a shorter timeframe such as five years.87

.

C. External requirements as fortress warning

The GATS does not distinguish between CUs and FTAs. But GATS Article V EIA looks like

a FTA since there is no obligation on its Members to adopt a common external tariff.88

CEMAC currently features a CU meanwhile a GATS EIA does not necessarily have to be

one. This would mean that in achieving liberalisation among them, individual CEMAC states

would keep their respective services market at committed level with the WTO.89

That

distinction between FTA and CU notwithstanding, Article V:4 GATS provides that the

83

Committee on Regional Trade Agreement, Examination of the North American Free Trade Agreement, Note

of the Meeting of 24 February 1997, WT/REG4/M/4, cited in Van den Bossche (2008), supra note 74, p. 712,

footnote 399. 84

Cottier and Molinuevo (2008), supra note 22, at 136 85

Ibid. 86

Van den Bossche, supra note 74, p. 712. 87

Stephenson (2000), supra note 5, p. 519, footnote 8 citing a proposal made by Japan at a meeting of the WTO

Council for Trade in Services in April 1999 where it is required of an EIA to eliminate discriminations in sectors

covered within five years after the entry into force of that agreement. GATT Article XXIV discussions have also

in some occasions suggested longer timeframes of say 12 years as practice shows (like in the case of the FTA

between the United States and Chile). See L. Bartels, ‘Interim Agreements’ under Article XXIV GATT, 8 World

Trade Review, no. 2 (2009), 339-350, at 346-348. See also Gathii (2011), supra note 4, p. 86 et seq where the

author argues inter alia for longer timeframe for developing countries, especially pursuant to the Enabling

Clause. 88

See Matsushita, Schoenbaum and Mavroidis (2006), supra note 29, p. 578 who argue that GATS PTA borrows

features of FTAs even though the term is not used as such. 89

Commitments are inscribed in each WTO Member’s GATS Schedule of Specific Commitments. This

Schedule is a document where a Member binds itself not to impose any new measures that would restrict entry

into the market beyond the specified levels of market access and national treatment in the document.

15

members to an EIA must not raise the level of barriers applicable to outsiders.90

In other

words, liberalisation must not be achieved at the expense of others.91

Although in assessing the extent to which this barrier has been raised vis-à-vis third

parties recourse has to be made to the level of barriers applicable prior to the conclusion of the

agreement, and in a sector/subsector by sector/subsector basis, difficulties stemming from the

almost impossibility to compute the “overall level of barriers” render the translation of this

requirement into practice hard to achieve. This is simply because barriers to trade in services

are for the most part not of a quantitative but of a qualitative nature.92

But since the provision

makes references to measures “applicable”, and not those “effectively applied”, what this

requirement means in reality is that Members of the EIA should not raise with respect to third

WTO Members barriers above their multilateral commitment.93

Additionally, it would mean

that parties of an EIA should not reduce the level of trade in services in any sector or

subsector after conclusion of the agreement,94

or that they should not reduce the level or

growth of trade in any sector or subsector below a historical trend.95

This in reality would be

translated into a prohibition of using liberalisation in the accounting services to balance

protection in legal services.

Article V:5 on the other hand sets compensatory adjustments in favour of affected parties

when the conclusion, enlargement or any significant modification of an EIA leads to the

withdrawal or modification of one of its Members’ specific commitments. An incumbent to

the PTA shall only proceed with the modification or withdrawal of concession – often to

comply with the EIA rules – upon having given a 90 days’ advance notice to the Council for

Trade in Services (CTS), after which notice the procedures of Article XXI of the GATS

(Modification of Schedules) in its paragraphs 2, 3 and 4 shall take effect. In all, any

introduction of measures contrary to one’s specific commitments at multilateral level while

joining an EIA triggers an immediate renegotiation process of those commitments as provided

for by Article XXI of the GATS.96

What Article V:5 stands out to say is that participation of a

WTO Member into an EIA shall not, with respect to other Members with which the

incumbent has concluded a contract prior to joining the scheme, be prejudicial. In this sense,

Article V:5 is simply lex specialis to Article V:4 because the former is an elaboration of the

latter.97

This is however without prejudice to any WTO Member, whether participating or not

90

Article V:4 states in relevant part that an EIA “shall in no respect of any Member outside the agreement raise

the overall level of barriers to trade in services within the respective sectors or subsectors compared to the level

applicable prior to such an agreement”. 91

Cottier (1996), supra note 27, at 159. 92

Stephenson (2000), supra note 5, p.519. 93

Cottier and Molinuevo (2008), supra note 22, p. 144. 94

Stephenson (2000), supra note 5, p.519. 95

Ibid. See also Van den Bossche (2008), supra note 74, p. 712. 96

Cottier and Molinuevo (2008), supra note 19, p. 145. Article XXI of the GATS provides for situations where a

WTO Member intends to modify or withdraw a commitment in its Schedule. Paragraphs 2, 3 and 4 of that

provision essentially deal respectively with “compensatory adjustments” in favour of affected Member(s),

arbitration proceedings in case no agreement has been reached between the modifying Member and the affected

one(s), and “retaliation” by the affected Member(s) if the modifying Member does not comply with the findings

of the arbitration. 97

Matsushita, Schoenbaum and Mavroidis (2006), supra note 29, p. 579. The legal maxim of statutory

interpretation “lex specialis derogat lege generali” (literally meaning that a “special law prevails over general

laws”), applies in situations of conflicts and may also give instructions on what a general rule requires in the case

at hand. Article V:4 and Article V:5 of the GATS are therefore governed by this general/special relationship. On

the study of “lex specialis” in international law, see for instance M. Koskenniemi, Fragmentation of

International Law: Difficulties Arising From the Diversification and Expansion of International Law, Report of

16

in an EIA, to introduce new restrictions on market access or national treatment in sectors and

modes of supply where specific commitments were not made.98

D. Notification

As a matter of procedure, conclusion of a GATS EIA must be promptly notified to the WTO

CTS accompanied by any relevant information that might be requested by the Council. This is

what comes out from the provision of GATS Article V:7. What then constitutes “prompt

notification” has been suggested to be notification occurring “no later than directly following

the parties’ ratification of the EIA, and before the application of preferential treatment

between the parties”. Hence, notification must be done prior to the implementation of

obligations and commitments pertaining to the EIA.99

In other words, notification to the CTS

is a prospective action of the potential EIA.

The purpose of notification as it stems from the wordings of Article V:7(a) is to allow the

CTS to examine its consistency with the provision of GATS V, hence its compatibility with

the multilateral rules. There is a possibility for the CTS to set up a working party to examine

the consistency of the agreement, but it is not required to do so. When the CTS elects to

establish that working party, EIAs are naturally referred to the Committee on Regional Trade

Agreements (CRTA) for examination. However, there has been no instance where a PTA was

ruled outright, if at all, to be WTO-inconsistent in the CRTA process.100

In fact, following the

creation of the Transparency Mechanism for Regional Trade Agreements “the multilateral

review has been narrowed down to a mere exercise in transparency”, and no such CRTA

report exists to date on the GATS PTAs consistency.101

In other words, the PTA review

mechanism has moved progressively from an ex ante to an ex post exercise, mainly due to the

fact that PTAs are often notified only after their conclusion and already operational.102

The application of Article V:7 GATS is also rendered difficult owing to the failure to

notify the services component of preferential agreements burgeoning in many parts of the

world. Many of these schemes by behaving like this fail to seize opportunities to deviate from

MFN in legality, especially when “flexibility” (for developing countries) is on the menu.

Others simply elect to operate in a state of illegality.103

Because review by the CRTA does not

always yield satisfactory results due to the “transparency exercise” as already pointed out,

existing PTAs also benefit from the fact that other WTO Members are reluctant to challenge

their functioning before the Dispute Settlement Body (DSB). If they were to, chances are that

the Study Group on Fragmentation of the International Law Commission, UN. Doc. A/CN.4/L.682 (13 April

2006), available at: <http://untreaty.un.org/ilc/documentation/english/a_cn4_l682.pdf>. 98

Cottier and Molinuevo (2008), supra note 22, p.146. It is again the place to note the vacuum in case no

commitment at all has been made at multilateral level. In such a scenario, Article V:5 is of less significance, if at

all. See also the text to note 68 supra. 99

See the Transparency Mechanism for Regional Trade Agreements Decision, WT/L/671, para. 3 of 18

December 2006, cited in Cottier and Molinuevo (2008), supra note 22, p. 149. 100

This has much to do with the unanimous decision method required within the CRTA to arrive (where, it is

important to note, incumbents will also be voting) as well as risk-averse Panels’ strategies. On PTAs’ review

mechanisms, see P. Mavroidis, Trade in Goods: the GATT and the Other Agreements Regulating Trade in

Goods, (Oxford: Oxford University Press, 2007), pp. 148-178. 101

The Transparency Mechanism for Regional Trade Agreements (of 14 December 2006) was meant to be a

“complement” to the existing arsenal (namely Article XXIV GATT, Article V GATS, Understanding on the

Interpretation of Article XXIV GATT, Decision on the establishment of the Committee on Regional Trade

Agreements), but ended up being a “substitute” to it. On this score, see Mavroidis (2010), supra note 9, at 1149.

See also Cottier and Molinuevo (2008), supra note 22, p. 150. 102

Mavroidis (2006), supra note 21, at 204 103

Stephenson (2000), supra note 5, p. 520.

17

it would result into the undesirable dispute battles.104

Risks for governments’ application of

preferential rules being challenged by a third WTO Member are however even greater if a

PTA was to be held illegal ex post. This would entail for companies losing the preference

scheme under which they have been conducting their business, hence losing their competitive

advantage.105

Although it is rather doubtful that a WTO Panel would order a straightforward dissolution

of a PTA for failure to comply with the rules of procedure, precaution would dictate for an

early notification to avoid the domino effect ensuing from a delayed notification. A strong

dispute settlement mechanism is further a way to balance between the proliferation of PTAs

and the need to ensure compliance with the multilateral discipline to avoid diversion of trade.

And since governments do not want confidence placed in them by firms to be shaken by

disputes that would nullify concessions under which they have been operating, this could

serve as a deterrent for those PTAs willing to operate inconsistently with the multilateral

rules. Practice seems to be at odd with this ideal nevertheless.

E. The Standard of Review

Article V:2 reads:

In evaluating whether the conditions under paragraph 1(b) are met, consideration may

be given to the relationship of the agreement to a wider process of economic

integration or trade liberalisation among the members concerned.

In order to decide whether the requirement of absence or elimination of “substantially all

discrimination” has been satisfied, the CTS (or based on the report of the CRTA) is allowed to

take into account the contribution of the notified EIA to the “wider process of economic

integration” between the members of the said EIA. In this vein, elements having to do with

the overall development policy of participating countries to the EIA may be relevant in

reviewing a notified scheme.106

This concept of “wider process of economic integration” has

been said in trade in goods to refer to the liberalisation pursuant to Article XXIV:5 GATT

1994.107

But it seems not to provide in a clear manner how to consider the relationship

between integration in goods and in services together. GATS Article V:1 as we have seen

requires intra-trade to be “substantially liberalised” among and between EIA’s members by

104

Despite the strength of WTO judicial mechanism, disputes do not abound. There exists what has been

described as a “mutual deterrence scenario”: Mavroidis (2006), supra note 21, at 211. This is the case even

though it is clear that the party invoking the application of the provision related to the establishment of the PTA

(i.e. the incumbent) bears the burden of showing that its scheme meets such requirement. It means that the

challenging party (i.e. the outsider), in principle, bears no risk (if not that of seeing its own PTA facing

challenges too). As Mavroidis puts it, motivations are possibly guided by the fact that “outsiders of today might

be the incumbents of tomorrow”. As such, they gain from not challenging today what they might be benefitting

from when they decide tomorrow to join the PTA, avoiding to “face the music [they] helped to compose”.

Moreover, it is not excluded that challenge on a PTA provision might lead to retaliation on a complete different

provision. To this risk-averse conduct, the author associates the often very high litigation costs (for sometimes

countries with no diplomatic presence in Geneva). See ample discussion by Mavroidis (2007), supra note 100,

pp. 171-177. 105

Cottier (1996), supra note 27, at 161. 106

Matsushita, Schoenbaum and Mavroidis (2006), supra note 29, p. 581. 107

That is, the external requirement not to impose overall barriers to outsiders to the CU and not to maintain

those of the members of the FTA, higher than those in place prior to the conclusion of the agreement. See the

Committee of Regional Trade Agreements, Synopsis of Systemic Issues Related to Regional Trade Agreements,

Note of the Secretariat, WT/REG/W/37 (2 March 2002), para. 85; and, Committee on Regional Trade

Agreements, Systemic Issues Arising from Article V of the GATS, Communication from Hong Kong, China,

WT/REG/W/34 (19 February 1999), para. 11, both cited in Cottier and Molinuevo (2008), supra note 22, p. 139.

18

the removal of “substantially all discrimination, and not vis-à-vis barriers to third-parties.

Article V:4 GATS – external requirement – is not subject to Article V:2 review. This is

understandable because members of a GATS EIA are not required to adopt a common

external policy.

Furthermore, the fact that consideration “may be given” to participation in a wider

integration process in both goods and services suggests that it is not a mandatory requirement

to review compliance of a notified EIA with the multilateral rules. It is instead an optional

consideration which, as has been argued, may lead to the CTS disregarding the relationship of

the services component with goods liberalisation altogether, or simply subjecting the

consideration of relationship between integration in goods and services to completely

different criteria (the level of development of parties for instance).108

This also fairly implies

that an EIA is not required to cover goods and services simultaneously, at least, as has been

put, on “formal” grounds absent a “specific obligation to that effect.”109

Also, the language

used here – that is its non-compelling character – has been attributed to the fact that the

CRTA’s review process is now one focused on transparency only, and, as such, does not

provide any practical effect of this standard of review provision.110

Although unlikely,

affected WTO Members (incumbents and outsiders together) are nonetheless not prevented

from lodging a complaint for non-conformity of an EIA with the “wider process of economic

integration” before a WTO judicial body.111

Despite these doubts, it would not be absurd to

see that a GATT-consistent PTA is judged GATS-consistent if liberalisation in services

among Members is not quite complete yet, since Article V:2 does not entirely rule out this

possibility. The immediate question that may follow, is whether Article V:2 could be

combined with that of “flexibility” under Article V:3 when reviewing south-south services

EIAs.112

Although a bit tangential, review of GATS EIAs could also be altered by the GATS

Annex to Article II Exemptions.113

In fact Article II:2 provides for the possibility for WTO

Members to deviate from MFN if they inscribe these inconsistent measures in their MFN

Exemptions List, provided, further, certain conditions are satisfied. It provides for justification

for giving more favourable treatment to the Member(s) specified in the List.114

Meant not to

last for more than ten years, its legal value is put in doubt since some Members have ascribed

a permanent status of their MFN exemptions in their Lists. The measure is available to new

Members that later decide to join the WTO. This raises a problem of compatibility of this

MFN exemption and the obligation of a Member in a EIA vis-à-vis the multilateral discipline.

An acceding country can make use of its once-and-for-all opportunity to inscribe in its List

108

Cottier and Molinuevo (2008), supra note 22, p. 140. The level of development, as we will see, is also a

criterion relevant in granting flexibility under Article V:3 GATS. 109

Matsushita, Schoenbaum and Mavroidis (2006), supra note 29, p. 581. 110

See text to note 101 supra. 111

It seems unlikely to happen because an incumbent would rarely want to help burst a PTA he helped to build

even if some Members to the agreement are not complying with their obligations. Yet, it is not excluded

completely. Whether the incumbent is in possession of a valid claim is another issue altogether. As for the

outsider, it is always possible, but again unlikely on the grounds that the EIA has breached the legal discipline of

its internal requirement (Article V:1 GATS) (which would probably have been dealt with by the CTS/CRTA, or

simply because it is a bad move to adopt since the less trade liberalisation exists among PTA members, the less

the risk of trade diversion, hence more gains potentials for outsiders). Still, the outsider could be tempted to

challenge the EIA on other grounds including, but not limited to, Article V:4 GATS. However, recall our

discussion supra note 124, and see “Article V:6” GATS infra. 112

On “flexibility”, see Section F infra. 113

Stephenson (2000), supra note 5, pp. 525-526. 114

WTO (1999), supra note 48, p.165, footnote 427.

19

some sectors that are required in order for the EIA to have “substantial sectoral coverage” in

the terms of Article V:1.

Another review hurdle linked to the reluctance to notify RTAs has to do with multiple and

overlapping memberships. This spaghetti bowl effect renders monitoring rather difficult.

F. Flexibility for Developing Countries’ Economic Integration Agreements

There is no GATS strict equivalent of the “Enabling Clause” as is the case in the sphere of

goods.115

When entering into a preferential services trade scheme, developing countries are

subject to the same discipline of Article V like developed countries. Multilateral review of

GATS EIAs is looser than in the goods context though. And it is even looser when developing

countries are involved. This is what emanates from the provision of Article V:3 GATS.

Article V:3(a) provides that:

“Where developing countries are parties to an agreement of the type referred to in

paragraph 1, flexibility shall be provided for regarding the conditions set out in

paragraph 1, particularly with reference to subparagraph (b) thereof, in accordance

with the level of development of the countries concerned, both overall and individual

sectors and subsectors.”

This paragraph requires that the standards applied with regard to ordinary EIAs be

disregarded in favour of a more relaxed one when developing countries are implicated in the

formation of such a bloc. Flexibility, however, is to be granted in accordance with the level of

development of the countries concerned. This implies that an EIA can as well feature

members with different levels of development, more precisely a developed country and a

developing country. This would further mean that all parties to the agreement are not

necessarily concerned by the flexibility, and that Article V:3 is not limited in the scope of

PTAs concluded among developing countries.116

In the case the agreement is between

developed and developing countries, flexibility will concern the reciprocity requirement of

“substantially all discrimination” elimination of Article V:1(b).117

This would entail that the

degree for granting national treatment would vary and be less constraining for the developing

country party.

Since the level of development of the countries concerned is to be taken into consideration

both the state and the prospects of competitiveness of the economy as a whole as well as those

of particular services sectors should be the basis of assessment.118

So, despite Article V:1(b)

discipline, flexibility may allow for lesser commitment in less competitive sectors, or full

exemption of these sectors and sub-sectors altogether or again allow for a longer transition

period in order to extent the “reasonable timeframe” (for liberalisation) of developing

countries.119

This interpretation sounds much as an “infant-industry” safeguard-type of

argument in GATT. Although GATS does not have a provision akin to this in the realm of

service yet, it might hence develop quite well from EIAs. It would be sound, though

debatable, to hold the same line of argument when the level of development is different

between two developing countries as well, especially when the performance in international

trade plus the degree of commitment at multilateral level are blatantly asymmetrical. This

115

Recall our discussion supra note 13 on LDCs services waiver. 116

Cottier and Molinuevo (2008), supra note 22, pp.140-141. 117

Ibid, p.141. 118

Ibid. 119

Ibid.

20

could therefore counter the difficulty relating to “self-election” as developing countries at the

WTO since this status remains until today uneasy to grasp.

What about asymmetry in development levels between CEMAC countries where intra-

trade in goods is not that high? CEMAC’s membership is composed of Equatorial Guinea

(classified by the United Nations as a Least Developed Country “LDC”)120

and not yet a

WTO Member, Central African Republic (CAR) and Chad (LDCs also), Cameroon, Congo

and Gabon (being “developing countries”). Chad and CAR are also land-locked countries.

What weight should be given to these factors while accessing the degree of trade liberalisation

among parties and the extent to which they are bound to grant national treatment among

themselves? If the deciding factor should be the degree of international competitiveness and

commitments it might be expected of Cameroon and Gabon to be more open than Chad or

CAR for instance. This poses another problem; that of the level of MFN the CAR is allowed

to grant to the other parties of the agreement should it decide to open its market more to Chad

with which it shares a “comparable level of development” and not to Gabon with which it

does not. The readings of Article V:3 lit. a does not quite provide a clear answer to this

question and we are tempted to argue that this will not be accepted even though allowing it

might not render the agreement per se illegal. But all will be a matter of scheduling and

whether a negative or a positive approach is adopted.

Article V:3(b) on the other hand reads:

Notwithstanding paragraph 6, in the case of an agreement of the type referred to in

paragraph 1 involving only developing countries, more favourable treatment may be

granted to juridical persons owned or controlled by natural persons of the parties to

such an agreement.

This provision talks about an agreement where all parties are developing countries. It governs

preferential rules of origin that parties to the agreement may choose to take advantage of or

not. When developing countries decide to implement it, the benefits accrue from the fact that

a firm is established and is conducting business in the territory of the parties to the agreement.

The qualification for benefitting of the flexibility is that the entity should be owned or

controlled by the individuals who are nationals of one of the parties to the EIA. This is

different in scope with Article V:6, as we will see, where the discipline relates to legal

persons controlled by natural persons of a third state. Article V:3(b) applies notwithstanding

Article V:6. Under this provision, EIAs members would lawfully discriminate in favour of an

Article V:3(b)-type firm even in the presence of an Article V:6-type firm.

It follows that preferential treatment to service suppliers owned or controlled by

individuals is restricted to citizens to EIAs concluded among developing countries.121

This in

turn amounts to a particular privilege granted to private ownership which departs from the

120

The UN uses the following criteria to identify LDCs: (i) a low-income criterion, based on a three-year

average estimate of the gross national income (GNI) per capita; (ii) a human resource weakness criterion,

involving a composite Human Assets Index (HAI) based on indicators of: (a) nutrition; (b) health; (c) education;

and (d) adult literacy; and (iii) an economic vulnerability criterion, involving a composite Economic

Vulnerability Index (EVI) based on indicators of: (a) the instability of agricultural production; (b) the instability

of exports of goods and services; (c) the economic importance of non-traditional activities (share of

manufacturing and modern services in GDP); (d) merchandise export concentration; and (e) the handicap of

economic smallness; and the percentage of population displaced by natural disasters. See the UN Office of the

High Representative for the Least Developed Countries, Landlocked Developing Countries and Small Island

Developing States, available at : <http://www.un.org/special-rep/ohrlls/ldc/ldc%20criteria.htm>, accessed 08

June 2013. 121

Cottier and Molinuevo (2008), supra note 22, p.142.

21

standard consideration of juridical persons whether owned or controlled by natural or juridical

persons.122

As such, it favours small and medium sized enterprises over enterprises of greater

importance that are subject to Article V:6, but does not at the same time mean that barriers to

outsiders should be unduly burdensome contra Article V:4.123

In summation, Article V:3 allows for broad flexibility when developing countries enter

into an EIA. Pursuant to that provision, the level of development of the country in question

must be taken into account while assessing the degree of flexibility its industries should enjoy.

It further allows developing countries to grant more favourable treatment to their services

providers provided it is controlled by their own nationals. However, paragraph (a) does not

necessarily imply paragraph (b). Rather, they can be combined depending on the status of the

parties. That is the reason why the flexibility of Article V:3(a) and the preferential rule of

origin of V:3(b) are “two independent and cumulative means to grant special and differential

treatment to developing countries in EAIs in services.”124

G. Foreign Direct Investments (FDIs) in EIAs or Liberal Rule of Origin: Article

V:6125

A company incorporated under the laws of a Member to the EIA and owned by individuals or

firms of another country must be granted the benefits of the EIA provided it engages into

substantive business operations in one of those countries.126

As the market becomes larger by

the conclusion of the EIA, opportunities for outsiders are also increased. Article V:6 thus

extends the benefits stemming from the conclusion of the EIA to the service suppliers of a

country not party to that scheme. In that sense, this is a revolutionary provision that

encourages FDIs in EIAs. The rationale behind it is probably to reduce trade-distorting effects

accruing from preferences inherent to EIAs with this extension to the benefit of any WTO

Member that happens to satisfy the conditions attached to it.127

RTAs members must therefore

be aware that their preferential scheme does not prevent third parties from benefitting from

the newly-formed larger market.

In concrete term, market access preferences and national treatment are henceforth granted

to service suppliers of a third country established in one country party to the agreement

without consideration as to whether they are owned or controlled by nationals of parties to

that agreement.128

A juridical person is defined under Article XXVIII (l) GATS as including

“any legal entity duly constituted or otherwise organized [sic] under applicable law, whether

for-profit or otherwise, and whether privately-owned or governmentally-owned, including any

corporation, trust, partnership, joint venture, sole proprietorship or association.” And since the

provision captures both duly constituted juridical persons (in the sense of incorporation) as

well as otherwise organised such legal entities, it surrounds entities like branches,

representative offices of foreign established corporations. This is because the various legal

procedures through which these “otherwise organised” entities have to pass for their

recognition are tantamount to “constitution”.129

In this sense, non-incorporated entities are

122

Ibid, p.143. 123

Ibid. 124

Ibid. 125

The phrase “liberal rule of origin” is borrowed from Cottier and Molinuevo (2008), supra note 22, p.146. 126

Article V:6 reads: “[a] service supplier of any Member that is a juridical person constituted under the laws of

a party to an agreement referred to in paragraph 1 shall be entitled to treatment granted under such agreement,

provided that it engages in substantive business operations in the territories of parties to such agreements.” 127

Cottier and Molinuevo (2008), supra note 22. p. 146 128

Ibid. 129

Ibid, p. 147.

22

embraced by the rule of origin since they are not expressis verbis excluded from this

provision.

A service supplier of a third WTO Member will take advantage of this provision and be

not discriminated against only on condition that it engages in “substantive business

operations” in the territory of a Member of the PTA. Substantive business operations have

been interpreted to mean “regular commercial activity” where “business operations” cover

activities like production, distribution, marketing, sale and delivery of a service.130

Cottier and Molinuevo further submit that this requirement can be viewed in two ways: a

wide view and a narrow one. Firstly, it could be said to mean that the “substantive business

operation” is to take place in the territory of either party, in which case the rules of origin are

cumulative.131

Under this reading a Chinese company for instance would have access to

Gabonese market and be granted national treatment simply by establishing a subsidiary in

Chad. Secondly, it could be interpreted to mean that benefit of the provision stemming from

substantive business operation would be accorded only in one territory especially where the

service supplier is established. This is a way to narrow down the scope of the provision to

actually catch only few of the service suppliers engaged in commercial activities in the

territory of one party to the EIA. These two interpretations are however correct in that

whether you take the first or the second, the result is to have a firm that qualifies as

conducting trade “in the territory of the parties” to claim for non-discrimination, and it does

qualify as such under each of the two grounds.

Article V:6, recall, applies without incidence to an EIA composed of only developing

countries. In the latter scenario, the relevant provision is Article V:3(b) where a “restricted”

rule of origin is applied militating for a more favourable treatment to service suppliers owned

by nationals of the parties to the agreement, taking precedence over Article V:6. Noteworthy

also is that the rule of origin as conceived here does not apply to natural persons under Mode

4, unfortunately.

H. Economic Integration Agreements’ Liberalisation Mechanisms

1. Scope of Liberalisation and Depth of Commitments

The scope also forms part of liberalisation strategy. EIAs either opt for universal sectoral

coverage where particularly sensitive sectors, such as air transport and audiovisual services,

are excluded.132

Liberalisation here can adopt a progressive method with an adjustment period

by which commitments should be implemented. It is also the case to feature in some EIAs

separate treatment of investments and movements of natural persons.133

Concerning the depth of commitments, their intensity affects the way countries reap

benefits of liberalisation at regional level. This is chiefly the case with the liberalisation of the

controversial Mode 4. It is therefore possible for members of an EIA to balance between

breadth (number of sectors covered) and depth (intensity of liberalisation of scheduled

commitments).

130

Article XXVIII (b). See the Negotiating Group on Rules, Compendium of Issues Related to Regional Trade

Agreements, Background Note by the Secretariat, Revision, TN/RL/W/8/Rev.1, paras 111-112 cited in Cottier

and Molinuevo (2008), supra note 22, p. 148, footnote 41. 131

Ibid, p.148. 132

UNCTAD (2007), supra note 3, p.7. 133

Ibid.

23

2. Liberalisation Modalities

Instantaneous free services trade in EIAs is rather chimerical to think of. Countries that

choose to liberalise services among them have done so in a GATS schedule of commitments

manner and do have the possibility to elect between two competing major methods: a positive

list approach and a negative list approach. It is also possible to adopt a “hybrid” method,

sometimes said to be reflected in the positive list approach itself.134

The positive list

resembles GATS’s scheduling mechanism – i.e. no sector and mode of supply is liberalised

unless expressly inscribed in the list of commitments – and provides much more flexibility

regarding the scope and pace of liberalisation. This “bottom-up” approach features

cautiousness on the part of parties to the EIA. The advantages it offers are that parties retain

the right not to disclose to their partners the remaining discriminatory measures and the

possibility to introduce new ones because they remain sovereign to undertake no

commitments.135

This method is prominent in agreements involving developing countries.

The positive list method is also described as hybrid in that it features “a voluntary,

positive, choice of sectors, sub-sectors and/or modes of supply in which governments are

willing to make binding commitments” [and] a negative list of non-conforming measures to

be retained in scheduled areas”.136

The negative list on the other hand operates in a “list it or lose it” fashion.137

That is, all

sectors and modes of supply are presumed to be liberalised and subject to the requirement of

non-discrimination within the trade bloc, unless a country expressly says the contrary by

listing sectors and modes to which restrictions remain and/or apply. Here, practice has shown

that parties to EIAs that opt for this approach usually insert in their “reservation lists” (that are

subject to periodic negotiations or consultations) existing restrictions and possibilities for

future ones so as to ensure transparency.138

Western Hemisphere-type agreements led by

NAFTA feature a negative listing method. Mexico’s participation in NAFTA has somewhat

extended this pattern in other PTAs in which it takes part.139

Negative list approach bears its own advantages and drawbacks. As far as drawbacks are

concerned, by not listing a particular sector countries automatically lose the right to introduce

discriminatory measures in future sectors including sector that are inexistent or simply not yet

regulated at the time the agreement enters into force.140

Another inconvenience of a lesser

magnitude is that it is burdensome to administer a scheme of this sort concluded under a

negative listing commitment and even more burdensome when countries involved lack

essential capacities like developing countries.141

Adjustment time –as a response – conceded

to some parties of that agreement to comply with their listing commitments might also blur

transparency from a private party standpoint.142

This fear should however be taken with a

pinch of salt when an agreement of this kind involves countries with a comparable level of

134

A. Mattoo and P. Sauvé, “Regionalism in Services Trade”, in A. Mattoo, R. Stern and G. Zanini (eds.), A

Handbook of International Trade in Services, (Oxford: Oxford University Press, 2007), p. 253. 135

Ibid. 136

Ibid. 137

Ibid. 138

Ibid, p. 254. 139

Ibid. In fact, it is the case with the Andean Community agreement that is said, however, to have adopted a

“somewhat different version of the negative list”: see Ibid, footnote 33. 140

Ibid., p. 256. 141

Ibid. 142

Ibid.

24

development and with comparable regulatory frameworks. Especially in certain South-South

PTAs (or North-North), it should be of less concern than in North-South PTAs.

Concerning the benefits of the negative scheduling approach, one can cite transparency in

general. This method of commitment signals the intention of one member not to rollback its

policy and allows potential traders and investors to find in the reservation lists a one-stop shop

of restrictions in foreign markets.143

V. EFFECTS OF REGIONAL TRADE LIBERALISATION IN SERVICES AND A

POTENTIAL CENTRAL AFRICA SERVICES TRADE BLOC

CEMAC was created in 1994 to replace the moribund UDEAC.144

Pursuant to Article I of

CEMAC Treaty of 1994, its objective is to promote a balanced development among the

members. Parties also intend to move from the existing state of cooperation among them to

that of a union capable of fulfilling the economic and monetary integration agenda.145

CEMAC is also open to other African countries sharing the same ideals – of solidarity,

freedoms and liberties, democracy, human rights and the rule of law – to join.146

This

provision means at least three things: firstly, CEMAC is not only an economic institution, but

also a political one; secondly, and probably more importantly, only an African country

sharing the same ideal can ask to join – which can explain partly why it qualified under the

“Enabling Clause”147

–; and thirdly, the aspiring African country must not necessarily be

contiguous to the present ones since the consideration is in terms of “ideals” and not that of

“proximity” (in geographical terms). This leads us to the early conclusion that an EIA among

these countries in light of the current treaty can only involve something among African States.

Although at the time of writing Equatorial Guinea as a CEMAC Party is not yet a WTO

Member, that does not affect our discussion since a PTA concluded between a WTO Member

and a non-Member still has to pass the test of Article V of the GATS in respect of the WTO

Member148

(i.e. Cameroon, Central African Republic, Chad, Congo and Gabon).

In order to achieve its objective, the Treaty established an economic union to complement

the pre-existing monetary union that was in place since the colonial period.149

Member States

empowered these two institutions to conduct the policies related to the elimination of all

obstacles to intra-trade within the region with the ultimate aim of achieving a common

market. The Economic Union of Central Africa (UEAC) is therefore the institution

empowered with the “economic integration” agenda.

143

Ibid. p. 255. 144

UDEAC stands for the Customs and Economic Union of Central Africa – a post-independence regional

grouping formed in 1964. Worth noting is that it’s during UDEAC years that the EC and ACP conventions

(Lomé I et seq) were entered into. Also, this period is marked by the creation of another, hence overlapping,

central African regional grouping, as a concretisation of the Lagos Plan of Action: ECCAS (in 1981). CEMAC

Treaty of 1994 was further revised in 2008. 145

See also Article 2 CEMAC Treaty (Revised). 146

Article 6 CEMAC and Article 55 CEMAC (Revised). This can qualify as a “semi-open” regionalism in that

any African State can join the scheme (kind of “open”), but limited to African only (sort of “closed”). 147

See WTO documents WT/COMTD/N/13 and WT/COMTD/24 of 29 September 2000. See for instance

paragraph 2(c) of the Enabling Clause that allows less developed Members to enter into PTAs not meeting the

strict requirements (of reciprocity, substantial all the trade, etc.) of Article XXIV GATT. 148

Cottier and Molinuevo (2008), supra note 22, p.130. 149

See the second intent of the Preamble of UEAC Convention.

25

A. CEMAC Common Market Ambitions

Calqued on the EU integration model, CEMAC RTA intends to create a Central Africa

Common Market based on the free movement of goods, services, capitals and persons.150

The

Convention governing the UEAC is the instrument that provides for the rules on elimination

of obstacles to trade in the CEMAC region. This text invites parties to a gradual and partial

conferral of their sovereignty in view of achieving the objective of regional integration.

Initially planned to be gradually achieved in three steps of five years each from the

entering into force of this Convention in 1999, parties were forced to reconsider their

ambitions after recording some delays in that sense. Against that backdrop a new Treaty was

signed in 2008 and another UEAC Convention was entered into to build upon the results

obtained from the first instrument. In this new vision, objectives of the economic union shall

be realised in two steps of three years each.151

Here, the common market which is to be real at

the end of the second phase will consist in the scheme going through Balassa stages of

economic integration.152

These ambitions, however, have yielded mitigated results. In fact, if CEMAC has

managed to form a monetary and a customs union and succeeded to harmonise competition

and business regulatory framework with a move towards macroeconomic convergence,153

it

has also not escaped documentation that CEMAC displays the lowest intraregional trade share

of less than 2 percent as compared to all other RTAs in Africa.154

A simplistic look at

statistics depicts CEMAC’s share of services exports also as one of the slowest to develop as

compared to other major RTAs in Africa. Exports of services for this sub-region are still

meaningless as compared to its ECOWAS counterpart not to mention its SADC or COMESA

during the same period. It illustrates that the value and share of services exports in these other

African RECs has more than quadrupled if not quintupled over time.155

Conversely, imports

have not ceased to increase and surpass the share of exports in this region. While the pattern is

similar in other parts of the world, the trade balance tends to equilibrium in some parts of

Africa. Again, cum grano salis, this is a rather non-conclusive statement because all factors

are not taken into account, especially the informal sector. These statistics should also be taken

with caution because of multiple and overlapping membership of some countries in these

RTAs – DRC (for the purpose of this paper) is an obvious example. Moreover, these statistics

account for international trade in general, and neither for interregional nor intra-regional

trade.

A question that flows naturally from these observations is whether CEMAC texts in their

present state are enough to be the driver of an emerging market where services form a

substantial part.

150

Per Article 2(c) UEAC Convention. 151

Article 3 UEAC Convention (Revised). 152

This is the provision of Article 13 of the UEAC Convention pertaining to the “common market”. These

Balassa’s five stages are the following: free trade areas, customs unions, common markets, economic unions and

complete economic integration as the final stage. See Bela Balassa, The Theory of Economic Integration

(Richard D Irwin Inc., 1961), p. 2. 153

UNCTAD, Economic Development in Africa Report 2009: Strengthening Regional Economic Integration for

Africa’s Development (Geneva: UNCTAD/ALDC/AFRICA/2009), p. 14. 154

Ibid. 155

See UNCTAD Handbook of Statistics 2011 (TD/STAT. 36, 18 December 2011) available at:

<http://unctad.org/en/Docs/tdstat36_en.pdf>, accessed 15 June 2012. Note ECCAS’ share of services export

over the same period is almost twice that of CEMAC thereby militating in favour of a “merger” of these blocs

for a meaningful impact.

26

B. Article V GATS and Central Africa Services Trade Agreement: Coexistence

or Deference?

1. Effects of Services PTAs

The characteristic of many preferential trade agreements mushrooming around the world has

been the increasing featuring of services trade component. This pattern is yet to be followed

in Africa and CEMAC sub-region in particular. Whether PTAs create or divert trade depends

on whether conclusion of a PTA is an end in itself or just a means to prepare participants to a

future multilaterally reciprocal trade, or as Hoekman and Sauvé argue, the issue is to a large

extent a function of the degree of discrimination against outsiders.156

Judging whether or not

such schemes are compatible with multilateralism, is pretty much dependent upon whether

regional agreements effectively lead to significant liberalisation and if they go substantially

beyond what is attainable in the multilateral setting.157

Consensus has not been found among

economists and the literature supporting either view is abundant.158

It is beyond the scope here

to reopen this endless debate. Nonetheless, services are different from goods and it remains

valid to see into their consequence.

Economic studies identify in the services area, more than in goods, greater potentials for

gains – static and dynamic – stemming from regionalism.159

In fact, Mattoo and Fink argue

that countries stand a better likelihood to benefit from preferential liberalisation of services

than they otherwise would if no (unilateral) liberalisation path is engaged.160

Thus, a country

will derive benefit from liberalising at least regionally (as opposed to multilaterally) instead

of opting for protectionism because barriers in services trade do not generate revenues and as

such do not increase global trade diversion costs. By so doing, the argument goes on, that

country’s economy takes advantage of the competition climate newly created and exploitation

of economies of scale coupled with the “learning by doing” effects the new economic

situation brings in. They warn, however, that other things equal, “non-preferential

liberalisation” is always better and yields larger welfare gains since it allows consumer to

choose among more competitive services suppliers coming into the market.161

Under these

circumstances non-competitive firms – be they domestic – simply have to exit the market.

As to the question whether there are circumstances where a country will be better off at

regional level than otherwise in a multilateral forum, Mattoo and Fink identify at least two

such circumstances, which they nevertheless cautioned are not services specific. Participants

in such a scheme are likely to gain at the expense of the rest of the world – unless outsiders

retaliate by concluding like agreements – because economic rents become concentrated in the

hands of oligopolists.162

Also, they maintain, it is more efficient to bargain and potentially

156

Hoekman and Sauvé (1994), supra note 28, at 314. 157

Ibid, at 313. 158

While proponents of PTAs on the one side of the spectrum argue that they bring about large trade creation

effects, the opponents on the other side consider PTAs harmful for global welfare. For the formers, see for

instance Baldwin (2006), supra note 2. See Also R. Baldwin and E. Seghezza, Are Trade Blocs Building or

Stumbling Blocks? New Evidence, CEPR Discussion Paper 6599 (2007). For the latter, see J. Bagwati, The

World Trading System at Risk (Princeton: Princeton University Press, 1991); P. Krugman, Is Bilateralism Bad?,

NBER Working Paper No. 2972 (May 1989). 159

UNCTAD (2007), supra note 3, p. 6. 160

A. Mattoo and C. Fink, Regional Agreements and Trade in Services: Policy Issues, World Bank Policy

Research Working Paper No. 2852 (June 2002), at 4. 161

Ibid., at 5. 162

Ibid.

27

lock-in policies among a subset of countries than it would otherwise be achievable on an

MFN basis at the multilateral level where countries are weary about free-riding.

The dynamic of gains is however qualified. Still according to Mattoo and Fink “the

sequence of liberalisation matters”, and countries could possibly lose in a long run if

multilateral liberalisation comes after preferential liberalisation.163

Vested interest of

incumbents firms that have been operating under preference might be difficult to reverse

when that country later decides to open its market on an MFN basis. This is basically due to

sunk costs necessary to enter that particular market. Adequate sequencing therefore holds the

key to contemplated long-term gains taking into account the characteristics of the sectors

involved.

2. Some Implications on CEMAC

The Convention governing the Central Africa Economic Union has clearly marked out the

reach of the agreement to encompass the freedom of movement of natural persons to seek

employment in a country to the RTA, that of establishment and the freedom to provide

services.164

For lack of clarity in framing the provisions, the reach of these freedoms is

however left to interpretation.

The freedoms of establishment and to provide services cover natural persons as well as

legal persons that are legally set up in the territory of a party to the agreement. In fact, they

grant independent individuals the right to settle in the territory of a Member State other than

theirs in order to engage therein in a non-remunerated activity, as well as in the acquisition,

constitution and management of companies.165

On these grounds, national treatment should be

extended to such service provider under the conditions that it is legally constituted under the

laws of a particular party, and has its headquarters or its principal place of business or of

management in the region. Furthermore, it can only benefit from the right of establishment on

condition that it demonstrates an effective and continuous economic link between its activity

and the economy of that particular Member.166

Freedom to provide services is full of discrimination nevertheless. In fact, domestic

regulations in many cases still operate in a way that nullifies the said objectives. For instance,

the law governing commercial activities in Cameroon subjects all foreigners that are not its

nationals to a prior authorisation and the granting of a licence to do so, regardless of whether

the person is from the territory of CEMAC or not, except in situations of mutual recognition

(agreement) in the sector involved.167

This also is the case of service suppliers already

established in the territory of another country of the agreement, which somehow impairs

investment. For example, that same law requires that at least 50 percent of the capital be

detained by Cameroonian nationals before applying for a licence, again except there is a

163

Ibid. 164

Article 27 UEAC Convention as read together with Article 13(d) (Revised). 165

E. Gnimpieba Tonnang, La libre circulation des personnes et des services en Afrique Centrale: entre

consécrations théoriques et hésitations politiques,71 Juridis Périodique (2007), at 94-95. 166

Ibid, at 95 footnote 95. This can appear as a limitation though because it allows the benefit of “national

treatment” only if a service provider is linked somehow with the economy of the country in which the treatment

is sought, and not in the territory of any other member of the community. This may call for a ‘mutual

recognition’-type of principle akin to what obtains in the European context (with the celebrated Cassis de Dijon

jurisprudence). 167

Ibid, at 96 footnote 103 on Article 9 of the Law n° 90/031 of 10 August 1990 regulating commercial activities

in Cameroon.

28

mutual agreement to that effect.168

These are just few of existing obstacles to trade in services

in particular. Theories and wishes are present but practice is absent.

Enjoying full legal personality under international law, CEMAC treaty and the convention

on UEAC that goes with it stress on a “coordination” and “harmonisation” methods to

integration. Although the sequence of liberalisation matters, the whole process has so far

appeared more as cooperation instead of integration. This is one of the reasons why it is weak

and could possibly be superseded by a much stronger instrument. No use (yet) has been made

of “flexibility” offered by GATS? Article V:3(b) for example?.

VI. CONCLUSIONS

This paper intended to test the proposition that central African countries should integrate their

economies in order to promote trade in services against the background of WTO Law on the

matter, and assessed the legal disciplines that would govern such integration. Such an

agreement to be GATS-consistent will have the stand the test of Article V, which is

nevertheless ‘relaxed’ when developing countries are members and which also accords

preferential rule of origin for the benefit of SMEs. Parties would further have to choose

between two tracks: either renegotiating a fresh services agreement (open to outsiders as the

case may be), or pass Community legislation in the field of services to give the Treaty

establishing the CEMAC and the UEAC Convention some teeth. The first track also paves the

way to the negotiations of services EPAs with the EU. Perceived to some extent as a threat to

African integration,169

EPA services agreements will certainly come under closer scrutiny

since developed countries will be involved.

However, grey areas in the interpretation of Article V of the GATS have not been

eliminated completely. It also does not come as a surprise given that GATS is a rather new

instrument and services PTAs are for the most part still in their infancy. In fact, Article V

discipline has not widely been tested yet. It was expected that the results of Doha negotiations

on RTA rules would finally clarify the extent of the disciplines of Article V GATS which

nevertheless remain vested in the Transparency Mechanism for Regional Trade Agreements

which in turn, as we have seen, has substituted itself to the existing arsenal narrowing down

review to a “mere exercise in transparency”.170

It has been submitted that there is a tendency at the WTO level by Members not to

challenge PTAs. We also pointed out that PTAs’ review is looser in services as compared to

goods PTAs, and more so when developing countries are implicated. This implies that,

despite the reluctance to challenge PTAs, even if the potential CEMAC service PTA (as an

EIA involving developing countries) were to face challenges, it would benefit from the

second shield commensurate to the status of its participants. Seen from this angle, failure to

take advantage of Article V GATS may be no strategy, but results from ignorance of potential

gains.

Countries that have liberalised their services markets have grown faster than those that

have not. This has proven true for developed economies and emerging developing countries.

168

Ibid, footnote 106. 169

See for instance ACORD, Economic Partnership Agreements: Jeopardizing a United Africa, Policy Briefing

Paper no.4 (June 2007), available at: <http://www.acordinternational.org/silo/files/economic-partnership-

agreements-jeopardising-a-united-africa.pdf>, accessed 16 January 2012. 170

See notes 101 and 102 supra.

29

Countries’ competitiveness has been established to be the function of their services industries.

Bad quality services leads to less competiveness and less choices to consumers. Many poor

countries are still to get on the train. This leads one to conclude that multilateralism is indeed

a jungle where only strong players impact on the flow of trade. Bilateralism and regionalism

sometimes are the best alternative for small economies, especially sub-Saharan African

markets, which are rather minuscule when taken individually. Hence, regionalism is not in

itself bad when conducted properly. But in order to yield benefits and attract investments, one

must be committed to the task and comply with the existing rules.

The scheduling mechanism of a central African EIA is also worth mentioning. GATS’s

“positive list” method, although laudable because it leaves individual and cautious states with

a bit of policy manoeuvres, will not be enough in this sub-regional setup. Against this

backdrop of “new regionalism” that militates for a wider and deeper integration, embarking in

a much stronger method is warranted. The objectives of CEMAC being to establish a

‘common market’, it is commendable to opt for a ‘negative list’ approach where all sectors

and modes of supply are liberalised unless expressly excluded.

Comments on the DDA and the desirability for developing countries to be more active are

worth making though. First, it shouldn’t be forgotten that reasons that militated for the

inclusion of SDT provisions for this group of countries are for the most part still present: their

tiny shops (i.e. their market) have not overnight turned into giant supermarkets; and if some

developing countries have gradually positioned themselves at a comparable level with

developed nations (to the extent that the call for their “graduation” could sound legitimate) it

is not however a uniform tendency across regions. Nevertheless, SDT provisions should not

be maintained just for their sake. The perverse nature of differential treatments has not

escaped documentation. Conditionality that has accompanied these “concessions” (local

content, structural adjustments) with sometimes less related trade issues (labour standards,

human rights), coupled with market distortions in the forms of farm and agriculture subsidies

in some parts of the West have rendered practice difficult for poor countries that relied on

crops as their main exports commodities. What was being given with the right hand was being

taken with the left. CEMAC sub-region is not alien to this scenario.

On the other hand, it is also evident that too much reliance on this variable geometry has

not helped built trade capacities, but increased addiction of the beneficiaries to aid. Inasmuch

as it is ideal to move from this pattern to one that would ensure participation of poorer nations

into the game of reciprocal trade, it is also undesirable to stifle their abilities to join forces to

be a stronger partner. If multilateralism is the end of the journey, regionalism in Africa should

be the gas station, not necessarily the parking place. A stop to fuel the engine, and not a stop

to have a rest, since, after all, globalisation and multilateralism have no rest.